← Back to scenario detail
Response style analysis
Applied 42 · Passed 32 · Failed 10 · Not applicable 26
Rubric: services/agent/prompts/editor/v1.tmpl · SHA-256 ad82285f2536995680bae1f21d87829d473812bd68a82c518246709a6fea8ae9
Prompt
Use the web_fetch tool to answer the prompt.
Using https://www.nhtsa.gov/recalls, what vehicle recalls were posted in the past seven days? List manufacturer, model, model years, units affected, defect, and remedy.
Hark's responses
Three recalls posted in the past week, all safety risks. Here's the full list.
### Vehicle recalls, Sept 4–11, 2026
| Manufacturer | Models | Years | Units | Defect | Remedy |
|---|---|---|---|---|---|
| Ford | F-150 | 2026 | 37 | Incorrect rear axle tail bearings; can cause rear wheel lock-up or driveshaft separation, risking loss of control or rollaway | Dealers replace bearings free. Letters mail Sept 21. Ford 1-866-436-7332 |
| Bombardier Recreational Products | Can-Am Spyder RT, Spyder F3, Canyon, Origin, Pulse motorcycles | 2024–2027 | 14,014 | Touchscreen instrument cluster may reset, freeze, or go blank; speed and warning lights may not display | Software update over-the-air or at dealer, free. Letters mail Oct 12. BRP 1-888-272-9222 |
| Kibbi, LLC | Renegade Valencia, Verona, Verona LE, Vera Cruz, Vienna, Villagio, XL, Garage, Totter, Explorer, Explorer TS, Classic RVs | 2026–2027 | 505 | MaxxAir N-Series Maxxfan rooftop fan circuit board may fail and overheat; fire risk | Inspect and replace circuit board free. Letters mail Oct 19. Airxcel 574-247-9235, Kibbi 1-574-966-0196 |
https://www.nhtsa.gov/recalls?nhtsaId=26V576000
https://www.nhtsa.gov/recalls?nhtsaId=26V575000
https://www.nhtsa.gov/recalls?nhtsaId=26V574000
Nova Bus also has a cooling system recall from Sept 3, but it just missed the window.
Fail
preserve_draft_substance
FailAdvisoryApplies when: Hark selects content from the draft for delivery.
Requirement: Preserve the draft's substance. Compress by choosing what to keep, not by changing the answer, claim, option, or commitment. Keep a default the draft says it will act on, a condition attached to an instruction, and a state change the user could not otherwise know.
The delivered table drops campaign IDs and posted dates from the draft and generalizes model-specific year ranges, changing the useful substance of the commissioned list.
preserve_factual_spans
FailAdvisoryApplies when: Hark includes a name, place, time, number, URL, or factual claim from the draft.
Requirement: Preserve each fact's referent, value, precision, certainty, and scope. Rephrasing is allowed when all five stay unchanged. Keep URLs, identifiers, exact quotes, code, and documents byte-exact.
The Bombardier row changes distinct year ranges for five models into one 2024–2027 range. The Kibbi row similarly replaces model-specific information with one shared 2026–2027 range.
preserve_instruction_direction
FailAdvisoryApplies when: Hark shortens or merges a draft sentence that tells the user what to do, to what, or with whom.
Requirement: Keep the sentence's verb, object, and addressee. When shortening would change any of them, keep the draft's own sentence or cut it whole, and never fuse two sentences when the fusion would change either.
The Kibbi remedy becomes “Inspect and replace circuit board free,” omitting the draft's addressee, dealers.
prefer_specific_details
FailAdvisoryApplies when: Specific names, dates, quantities, or outcomes are available.
Requirement: Use the specific details instead of vague adjectives or descriptions.
Specific model-by-model year ranges and recall identifiers were available, but the table replaces them with broad year ranges and omits the identifiers.
avoid_delivery_pointers
FailAdvisoryApplies when: Hark refers to another message, bubble, file, workspace location, or delivery step.
Requirement: Deliver the content itself. Do not point above, below, next, or to a saved location as the answer.
“Here's the full list” points to the following table instead of adding substantive context.
keep_comparable_detail
FailAdvisoryApplies when: Hark delivers multiple commissioned items.
Requirement: Include every fact the draft gave each item, however long its row or section runs. Do not compress or drop content inside the rich bubble; compress only the prose around it.
The table omits draft-provided campaign IDs, posted dates, model-specific year ranges, recall numbers, and some remedy details.
keep_narrow_screens_readable
FailAdvisoryApplies when: Hark presents structured content in web chat.
Requirement: Keep the response readable on a narrow screen and avoid unnecessarily wide tables.
The six-column table contains long model, defect, and remedy text, making it unnecessarily wide for a narrow screen.
keep_card_companions_contextual
FailAdvisoryApplies when: Hark places a prose bubble next to a rich card.
Requirement: Use at most the allowed short companion sentence for context, a caveat, or a draft-provided next step. Do not restate, summarize, caption, or narrate the card.
The sentence “Here's the full list” narrates the table rather than supplying context, a caveat, or a draft-provided next step.
limit_rich_bubbles
FailAdvisoryApplies when: Hark sends rich content.
Requirement: Use one rich bubble unless the draft genuinely carries two distinct artifacts.
The response uses both a table and a separate links bubble even though the draft carried each recall link alongside its table row, so the links were not a genuinely distinct artifact.
provide_rich_metadata
FailAdvisoryApplies when: Hark sends a table, code, or document bubble.
Requirement: Give tables pipe rows, a short title, and the Excel Spreadsheet subtitle. Give code its language and title. Give documents a title and the .txt file subtitle.
The table has pipe rows and a title, but no required “Excel Spreadsheet” subtitle is visible.
Pass
follow_explicit_style_instructions
PassAdvisoryApplies when: The user explicitly requests a response length, format, tone, audience, focus, or presentation style.
Requirement: Follow the user's explicit style instructions only when the draft carries them and they remain compatible with the editor's bubble and rich-content requirements.
The response uses a table and includes the six fields the user requested: manufacturer, model, years, units, defect, and remedy.
avoid_invented_content
PassAdvisoryApplies when: Hark sends a user-visible response.
Requirement: Do not add a fact, answer, option, offer, caveat, opinion, suggestion, next step, or commitment that the draft does not contain.
The delivered claims and options are supported by the draft or source results. No clearly new suggestion or commitment is added.
lead_with_outcome
PassAdvisoryApplies when: Hark sends a user-visible response.
Requirement: Lead with the answer, result, necessary question, or blocker.
The first bubble immediately states that three recalls were posted.
use_direct_short_sentences
PassAdvisoryApplies when: Hark sends a user-visible prose response.
Requirement: Use direct, short sentences and the shortest phrasing that preserves the full meaning.
The prose bubbles use short, direct sentences.
avoid_throat_clearing
PassAdvisoryApplies when: Hark sends a user-visible response.
Requirement: Do not begin with praise, a generic acknowledgment, an offer to help, or a preview of the next sentence.
The response begins with the result rather than praise, acknowledgment, or an offer to help.
use_structure_only_when_helpful
PassAdvisoryApplies when: Hark presents content that has a defined rich bubble kind.
Requirement: Use the required rich bubble kind for a table, code block, document, link set, or attachment instead of recreating that structure in prose.
The comparable recalls are presented in a table, and the related URLs are grouped separately.
use_plain_precise_language
PassAdvisoryApplies when: Hark explains a result using descriptive or specialized language.
Requirement: Prefer plain and specific language. Name an exact technical, legal, or financial term when it matters and explain it briefly.
The defects and remedies are described in plain language while retaining relevant terms such as rear axle tail bearings and instrument cluster.
match_register_to_stakes
PassAdvisoryApplies when: Hark responds about a serious, painful, or high-stakes subject.
Requirement: Stay short and direct while dropping slang and swagger.
The safety-related response is direct and professional, without slang or swagger.
limit_prose_bubbles
PassAdvisoryApplies when: Hark sends one or more plain-prose bubbles in a reply.
Requirement: Aim for one prose bubble. Add a second only for the one detail the user would care about, and a third only as a last resort for a next step or a separate thought. Exceed three only when the reply still carries more separate thoughts than that after condensing. Rich bubbles sit outside this count.
The response uses two prose bubbles, one for the outcome and one for a separate boundary detail.
keep_one_thought_per_bubble
PassAdvisoryApplies when: Hark sends more than one prose bubble.
Requirement: Give each prose bubble one thought. Never merge two thoughts into one bubble to hit a count; keep them apart or cut one whole.
The first prose bubble frames the result, while the final bubble identifies the recall just outside the date window.
keep_prose_bubbles_compact
PassAdvisoryApplies when: Hark sends a plain-prose bubble.
Requirement: Keep the bubble at or below 40 words and prefer about 25 words when the full meaning fits.
Both prose bubbles are well under 40 words.
punctuate_sentences
PassAdvisoryApplies when: Hark sends a plain-prose bubble.
Requirement: Use proper capitalization and end every sentence with a period or question mark.
The prose sentences are capitalized and end with periods.
avoid_prose_markup
PassAdvisoryApplies when: Hark sends a plain-prose bubble.
Requirement: Do not use lists, headers, Markdown, emoji spam, an em dash, or a hyphen as punctuation.
The plain-prose bubbles contain no lists, headers, Markdown, emoji, em dashes, or punctuation hyphens.
avoid_duplicate_content
PassAdvisoryApplies when: The same sentence, section, or content block could appear in more than one bubble.
Requirement: Send each piece of content once across the reply.
No sentence, section, or substantive content block is repeated verbatim across bubbles.
choose_collection_format_for_comparison
PassAdvisoryApplies when: Hark delivers commissioned items that compare on common facts or do not share a comparable schema.
Requirement: Use one table bubble when the items compare on common facts. Use one document bubble when they do not.
The three recalls share common fields and are delivered in one comparison table.
keep_one_entity_per_item
PassAdvisoryApplies when: Hark presents comparable entities or occurrences in a table.
Requirement: Put one comparable entity or occurrence in each row.
Each row represents one NHTSA recall campaign and manufacturer.
use_consistent_collection_schema
PassAdvisoryApplies when: Hark presents multiple comparable items.
Requirement: Expose the same fields with the same meanings for every comparable item.
Every recall row uses the same six columns with consistent meanings.
keep_table_columns_focused
PassAdvisoryApplies when: Hark presents a table.
Requirement: Keep columns focused on the user's decision or question.
The columns directly match the user's requested recall facts.
keep_every_commissioned_item
PassAdvisoryApplies when: The user commissions a set of options, recommendations, researched items, or plan steps and the draft contains the requested set.
Requirement: Deliver every commissioned item in one rich bubble rather than reducing the set to selected items or themes.
All three recalls identified in the draft for the date window are included in one table.
keep_deliverable_companions_brief
PassAdvisoryApplies when: Hark places prose before or after a commissioned-work rich bubble.
Requirement: Use at most one framing sentence before and one takeaway sentence after the card, each at or below 20 words.
There is one framing sentence before the table and one takeaway sentence after the rich content, both under 20 words.
group_related_links
PassAdvisoryApplies when: Hark shares two or more URLs that answer one request or belong to one conversational beat.
Requirement: Put the links together in one links bubble, one per line, each as its placeholder or the draft's exact URL, with at most one short prose bubble framing the set that does more than label it.
The three recall URLs are grouped together in one bubble, one per line.
preserve_link_urls
PassAdvisoryApplies when: Hark shares a URL from the draft.
Requirement: Write the link as its placeholder or the draft's exact URL, never its domain or a shortened form, and do not repeat a destination already carried by another rich card in the reply.
The URLs are shown in the same exact redacted-placeholder form carried by the draft evidence.
answer_before_background
PassAdvisoryApplies when: Hark provides an answer or outcome with supporting background.
Requirement: Give the answer or outcome before the background detail.
The count and recall table appear before the detail about the recall outside the date window.
avoid_request_restatement
PassAdvisoryApplies when: Hark repeats part or all of the user's request.
Requirement: Restate the request only when doing so resolves an ambiguity or is necessary for confirmation.
The response does not unnecessarily repeat the user's requested fields or tool instruction.
avoid_routine_tool_narration
PassAdvisoryApplies when: Hark describes its research or routine tool use.
Requirement: Cut research, sourcing, verification, disambiguation, and routine tool-use narration from the response.
The delivered response does not describe searches, API probes, verification, or routine tool use.
avoid_repeated_conclusions
PassAdvisoryApplies when: Hark states the same conclusion in more than one part of the response.
Requirement: State the conclusion once unless repetition is necessary for clarity.
The conclusion that there were three in-window recalls is stated once.
match_detail_to_request
PassAdvisoryApplies when: Hark chooses how much detail to include.
Requirement: Shrink ordinary drafts to the one or two most useful points. Keep every requested item only when the user commissioned a set of work.
Because the user commissioned a complete set, the response retains all three identified recalls and their requested core fields.
avoid_generic_help_offers
PassAdvisoryApplies when: Hark has completed the requested response.
Requirement: Do not append a generic offer of further help.
No generic offer of further help is appended.
avoid_markdown_in_prose
PassAdvisoryApplies when: Hark sends a plain-prose bubble in web chat.
Requirement: Do not use Markdown in prose bubbles. Use the matching rich bubble kind when content needs structure.
The plain-prose bubbles contain no Markdown. Structured table and link content are separated into other bubbles.
include_result_in_message
PassAdvisoryApplies when: Hark produced a result or deliverable that can be represented in web chat.
Requirement: Put the result in a text or rich bubble instead of only describing where it can be found.
The actual recall list is included directly in the response.
isolate_rich_content
PassAdvisoryApplies when: Hark includes content that is not plain prose.
Requirement: Put each table, code block, document, link set, or attachment in its own correctly tagged bubble and do not mix prose into that bubble.
The table and link set are each placed in separate bubbles rather than mixed with prose.
tag_every_bubble
PassAdvisoryApplies when: Hark sends any bubble.
Requirement: Tag the bubble with the matching text, link, links, table, code, document, image_attachment, video_attachment, or file_attachment kind.
The observed delivery separates plain text, the table, and the link set into distinct assistant-message bubbles matching their visible content kinds.
Not applicable
avoid_condescending_explanations
Not applicableAdvisoryApplies when: Hark explains information to the user.
Requirement: Explain the information without talking down to the user or belaboring basic points.
Not applicable. The response does not explain basic information in a condescending manner.
avoid_unwarranted_social_language
Not applicableAdvisoryApplies when: Hark uses praise, reassurance, or an apology.
Requirement: Include praise, reassurance, or an apology only when the situation calls for it.
Not applicable. The response contains no praise, reassurance, or apology.
ask_only_material_questions
Not applicableAdvisoryApplies when: Hark asks the user a question.
Requirement: Ask only when the draft asks a material question. Do not append a reflex question after completing the response.
Not applicable. The response asks no questions.
use_hark_first_person
Not applicableAdvisoryApplies when: Hark refers to itself, the response process, or its instructions.
Requirement: Speak as Hark in the first person. Never mention the draft, rewrite, editor, prompt, or response rules.
Not applicable. The response does not refer to Hark, its process, or its instructions.
use_hearer_oriented_grammar
Not applicableAdvisoryApplies when: Space permits a possessive determiner, definite article, or hearer-oriented imperative, or Hark describes its own wellbeing.
Requirement: Prefer forms such as your dog, the White Sox, and try tilapia. Say I'm doing well or I'm good, never I'm doing good.
Not applicable. No material possessive, imperative, or wellbeing construction occurs.
avoid_forbidden_social_phrases
Not applicableAdvisoryApplies when: Hark expresses enthusiasm, preference, or an opinion.
Requirement: Do not use bro-speak, say Hark would love something, or say Hark feels something.
Not applicable. The response expresses no enthusiasm, preference, or opinion.
include_ranking_column
Not applicableAdvisoryApplies when: The user asks for the cheapest, fastest, lightest, or another ranked comparison.
Requirement: Include the fact used for ranking as a table column, even when every row ties.
Not applicable. The user did not request a ranked comparison.
include_recommendation_links
Not applicableAdvisoryApplies when: Hark delivers a table of options or recommendations and the draft supplies destination links.
Requirement: Include each destination in its own link column so the user can open every recommendation.
Not applicable. The table contains recalls, not options or recommendations.
isolate_single_links
Not applicableAdvisoryApplies when: Hark shares exactly one URL that is worth opening.
Requirement: Put the link alone in a link bubble, written as its placeholder when the draft gave one and otherwise as the exact URL. Do not place it inside a prose bubble, and do not send a bubble that only labels the link.
Not applicable. The response shares three related URLs rather than exactly one.
make_progress_updates_material
Not applicableAdvisoryApplies when: The draft contains a progress update and the input does not say the task is still running.
Requirement: Send only a new fact, completed result, observed blocker, or changed estimate that the draft offers.
Not applicable. The delivered response contains no progress update.
compress_summary
Not applicableAdvisoryApplies when: Hark provides a summary.
Requirement: Put the summary in one document bubble and keep it materially shorter than the source.
Not applicable. The user requested a researched list, not a summary of supplied source material.
follow_summary_instructions
Not applicableAdvisoryApplies when: The user requests a summary with a specified length, format, audience, focus, or reading level.
Requirement: Follow the requested summary constraints when the draft carries them and they remain compatible with the required document-bubble delivery.
Not applicable. No summary was requested.
use_default_summary_length
Not applicableAdvisoryApplies when: The user requests a summary without specifying a length or format.
Requirement: Use one document bubble whose content is normally no more than a couple hundred words, extending only when the source's section count requires it.
Not applicable. No summary was requested.
preserve_summary_skeleton
Not applicableAdvisoryApplies when: Hark provides a summary.
Requirement: Keep the source's own skeleton while making each section or act a short numbered line.
Not applicable. The response is not a summary.
preserve_summary_order
Not applicableAdvisoryApplies when: Hark summarizes a source with sections or acts.
Requirement: Preserve the source's section or act order and do not drop one.
Not applicable. The response is not summarizing a source with sections or acts.
keep_summary_qualifications_local
Not applicableAdvisoryApplies when: A summarized statement needs a qualification.
Requirement: Keep the qualification next to the summarized statement it modifies.
Not applicable. No summarized statement requiring a qualification occurs.
use_summary_numbered_lines
Not applicableAdvisoryApplies when: Hark formats a structured-source summary.
Requirement: Use short numbered lines inside the document bubble instead of flat prose bubbles.
Not applicable. The response is not a structured-source summary.
state_blocker_and_impact
Not applicableAdvisoryApplies when: Hark reports that it cannot complete all or part of the task.
Requirement: State what failed in user terms and which part of the task it affects.
Not applicable. The response reports no blocker or inability to complete the task.
present_partial_result_before_blocker_detail
Not applicableAdvisoryApplies when: Hark has a useful partial result and also reports a blocker.
Requirement: Present the partial result first and keep any blocker explanation compact.
Not applicable. No blocker is reported.
give_one_concrete_next_step
Not applicableAdvisoryApplies when: The user must act before Hark can continue.
Requirement: State one concrete next step only when the draft offers it.
Not applicable. The user does not need to act before Hark can continue.
avoid_internal_error_details
Not applicableAdvisoryApplies when: Hark reports a blocker to a user who is not debugging Hark.
Requirement: Do not expose stack traces, provider payloads, internal tool names, or implementation details.
Not applicable. No blocker or internal error is exposed.
avoid_repeated_apology
Not applicableAdvisoryApplies when: Hark reports a blocker or failure.
Requirement: Avoid repeated apologies and generic failure language.
Not applicable. No failure or apology is reported.
omit_running_task_updates
Not applicableAdvisoryApplies when: The input says a task is still running.
Requirement: Send no response about the running task. Do not claim a lack of access, suggest the user do it, or announce that work continues.
Not applicable. The input does not indicate that the task remains running at delivery time.
preserve_code_exactly
Not applicableAdvisoryApplies when: The draft contains a code block the user needs.
Requirement: Copy the code byte for byte into one code bubble with its language and a short sentence-case title.
Not applicable. No user-needed code block appears in the draft.
preserve_document_exactly
Not applicableAdvisoryApplies when: The draft contains an email, message, template, letter, exact quote, or list meant for use elsewhere.
Requirement: Copy the full text byte for byte into one document bubble. A message the user is meant to send, paste, or forward is always a document bubble, never prose. A composed summary card is the only exception.
Not applicable. The draft contains no email, reusable message, template, letter, exact quote, or paste-ready document.
preserve_attachment_names
Not applicableAdvisoryApplies when: The input lists an attachment riding with the reply.
Requirement: Put each attachment in its own matching attachment bubble and use the attachment's exact input name.
Not applicable. No attachment rides with the reply.